Advanced Lane Finding¶
Import Packages¶
In [2]:
#importing some useful packages
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import numpy as np
import cv2
import glob
%matplotlib inline
Distortion Correction¶
Camera Calibration¶
In [3]:
#example for a single image
#reading in an image
image = cv2.imread('camera_cal/calibration2.jpg')
#printing out some stats and plotting
print('This image is:', type(image), 'with dimensions:', image.shape)
plt.imshow(image) # if you wanted to show a single color channel image called 'gray', for example, call as plt.imshow(gray, cmap='gray')
#Arrays to store object points and image points from all the images
plt.show()
objpoints = [] #3D points in real world space
imgpoints = [] #2D points in image plane
#Prepare object points, like(0,0,0), (1,0,0)...
objp = np.zeros((6*9, 3), np.float32)
objp[:,:2] = np.mgrid[0:9, 0:6].T.reshape(-1,2) #x, y coordinates
#Convert image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
#Find the chessboard corners
ret, corners = cv2.findChessboardCorners(gray, (9,6), None)
# If corners are found, add object points, image points
if ret == True:
imgpoints.append(corners)
objpoints.append(objp)
#draw and display the corners
img = cv2.drawChessboardCorners(image, (9,6), corners,ret)
plt.imshow(img)
In [4]:
objpoints = [] # 3d points in real world space
imgpoints = [] # 2d points in image plane.
#Prepare object points, like(0,0,0), (1,0,0)...
objp = np.zeros((6*9,3), np.float32)
objp[:,:2] = np.mgrid[0:9, 0:6].T.reshape(-1,2) #x, y coordinates
images = glob.glob('camera_cal/calibration*.jpg') # all images path
for i, filename in enumerate(images):
img = cv2.imread(filename)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# Find the chessboard corners
ret, corners = cv2.findChessboardCorners(gray, (9,6), None)
# If found, add object points, image points
if ret == True:
objpoints.append(objp)
imgpoints.append(corners)
# Draw and display the corners
f, (ax1, ax2) = plt.subplots(1, 2, figsize=(20,10))
ax1.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
ax1.set_title('Image'+str(i+1), fontsize=25)
cv2.drawChessboardCorners(img, (9,6), corners, ret)
ax2.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
ax2.set_title('show corners', fontsize=25)
Distortion Correction¶
In [5]:
# define undistort function
def undistort(img):
img_size = (img.shape[1], img.shape[0])
ret, mtx, dist, rvecs, tvecs = cv2.calibrateCamera(objpoints, imgpoints, img_size, None, None)
undst = cv2.undistort(img, mtx, dist, None, mtx)
return undst
In [7]:
images = glob.glob('test_images/*.jpg')
for image in images:
img = cv2.imread(image)
undst = undistort(img)
f, (ax1, ax2) = plt.subplots(1, 2, figsize=(12,6))
ax1.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
ax1.set_title('test image', fontsize=25)
ax2.imshow(cv2.cvtColor(undst, cv2.COLOR_BGR2RGB))
ax2.set_title('undistorted image', fontsize=25)
In [48]:
#single image
img = cv2.imread('test_images/straight_lines1.jpg')
#undistort image
undist = undistort(img)
undist, unpersp, Minv = per_transform(undist)
plt.imshow(undist)
plt.show()
#graysclae
gray = cv2.cvtColor(undist, cv2.COLOR_BGR2GRAY)
#HLS color space
hls = cv2.cvtColor(undist, cv2.COLOR_BGR2HLS)
s_channel = hls[:,:,2]
# apply gradient threshold on the horizontal gradient
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0) # Take the derivative in x
abs_sobelx = np.absolute(sobelx) # Absolute x derivative to accentuate lines away from horizontal
scaled_sobel = np.uint8(255*abs_sobelx/np.max(abs_sobelx))
# Threshold color channel
thresh_min = 20
thresh_max = 100
sxbinary = np.zeros_like(scaled_sobel)
sxbinary[(scaled_sobel >= thresh_min) & (scaled_sobel <= thresh_max)] = 1
plt.imshow(sxbinary)
plt.show()
# Threshold color channel
s_thresh_min = 90
s_thresh_max = 255
s_binary = np.zeros_like(s_channel)
s_binary[(s_channel >= s_thresh_min) & (s_channel <= s_thresh_max)] = 1
plt.imshow(s_binary)
plt.show()
color_binary = np.dstack(( np.zeros_like(sxbinary), sxbinary, s_binary))
plt.imshow(color_binary)
plt.show()
# Combine the two binary thresholds
combined_binary = np.zeros_like(sxbinary)
combined_binary[(s_binary == 1) | (sxbinary == 1)] = 1
plt.imshow(combined_binary)
plt.show()
#region of interst
mask = np.zeros_like(combined_binary)
length = mask.shape[1]
height = mask.shape[0]
vertices = np.array([[(0,height),(length/2, height/2), (length-1, height-1)]], dtype=np.int32)
cv2.fillPoly(mask, vertices, 1)
masked_image = cv2.bitwise_and(combined_binary, mask)
plt.imshow(masked_image)
plt.show()
In [35]:
images = glob.glob('test_images/*.jpg')
for image in images:
img = cv2.imread(image)
masked_image, combined_binary, color_binary, s_binary, sxbinary = threshold(img)
f, (ax1, ax2) = plt.subplots(1, 2, figsize=(12,6))
ax1.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
ax1.set_title('test image', fontsize=25)
ax2.imshow(masked_image)
ax2.set_title('thresholded image', fontsize=25)
In [ ]:
Perspective Transform¶
In [6]:
#define perspective transform function
def per_transform(undist):
img_size = (undist.shape[1], undist.shape[0])
src = np.float32([[220,720],[1110,720],[722,470],[570,470]])
dst = np.float32([[320,720],[920,720],[920,1],[320,1]])
M = cv2.getPerspectiveTransform(src, dst)
Minv = cv2.getPerspectiveTransform(dst, src)
warped = cv2.warpPerspective(img, M, img_size, flags=cv2.INTER_LINEAR)
unpersp = cv2.warpPerspective(warped, Minv, img_size, flags=cv2.INTER_LINEAR)
return warped, unpersp, Minv
In [53]:
images = glob.glob('test_images/*.jpg')
for image in images:
img = cv2.imread(image)
masked_image, combined_binary, color_binary, s_binary, sxbinary = threshold(img)
warped, unpersp, Minv = per_transform(masked_image)
f, (ax1, ax2) = plt.subplots(1, 2, figsize=(12,6))
ax1.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
ax1.set_title('test image', fontsize=25)
ax2.imshow(cv2.cvtColor(warped, cv2.COLOR_BGR2RGB))
ax2.set_title('warped image', fontsize=25)
Color & Gradient Threshold¶
In [7]:
#Define a function for lane detection
def threshold(img):
#undistort image
undist = undistort(img)
#per-transform
wraped, unpersp, Minv = per_transform(undist)
#graysclae
gray = cv2.cvtColor(wraped, cv2.COLOR_BGR2GRAY)
#HLS color space
hls = cv2.cvtColor(wraped, cv2.COLOR_BGR2HLS)
s_channel = hls[:,:,2]
# apply gradient threshold on the horizontal gradient
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0) # Take the derivative in x
abs_sobelx = np.absolute(sobelx) # Absolute x derivative to accentuate lines away from horizontal
scaled_sobel = np.uint8(255*abs_sobelx/np.max(abs_sobelx))
# Threshold color channel
thresh_min = 20
thresh_max = 100
sxbinary = np.zeros_like(scaled_sobel)
sxbinary[(scaled_sobel >= thresh_min) & (scaled_sobel <= thresh_max)] = 1
# Threshold color channel
s_thresh_min = 90
s_thresh_max = 255
s_binary = np.zeros_like(s_channel)
s_binary[(s_channel >= s_thresh_min) & (s_channel <= s_thresh_max)] = 1
# Stack each channel to view their individual contributions in green and blue respectively
# This returns a stack of the two binary images, whose components you can see as different colors
color_binary = np.dstack(( np.zeros_like(sxbinary), sxbinary, s_binary))
# Combine the two binary thresholds
combined_binary = np.zeros_like(sxbinary)
combined_binary[(s_binary == 1) | (sxbinary == 1)] = 1
#region of interst
mask = np.zeros_like(combined_binary)
length = mask.shape[1]
height = mask.shape[0]
vertices = np.array([[(0,height),(length/2, height/2), (length-1, height-1)]], dtype=np.int32)
cv2.fillPoly(mask, vertices, 1)
masked_image = cv2.bitwise_and(combined_binary, mask)
return masked_image, combined_binary, color_binary, s_binary, sxbinary, Minv
In [50]:
images = glob.glob('test_images/*.jpg')
for image in images:
img = cv2.imread(image)
masked_image, combined_binary, color_binary, s_binary, sxbinary = threshold(img)
f, (ax1, ax2) = plt.subplots(1, 2, figsize=(12,6))
ax1.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
ax1.set_title('test image', fontsize=25)
ax2.imshow(combined_binary)
ax2.set_title('thresholded image', fontsize=25)
Histogram¶
In [59]:
img = cv2.imread("test_images/straight_lines1.jpg")
warped, combined_binary, color_binary, s_binary, sxbinary = threshold(img)
histogram = np.sum(warped[warped.shape[0]//2:,:], axis=0)
# Peak in the first half indicates the likely position of the left lane
half_width = np.int(histogram.shape[0]/2)
leftx_base = np.argmax(histogram[:half_width])
# Peak in the second half indicates the likely position of the right lane
rightx_base = np.argmax(histogram[half_width:]) + half_width
print(leftx_base, rightx_base)
plt.plot(histogram)
plt.xlabel("Pixel position")
plt.ylabel("counts")
Out[59]:
Sliding Windows¶
In [86]:
img = cv2.imread("test_images/straight_lines1.jpg")
masked, binary_warped, color_binary, s_binary, sxbinary = threshold(img)
# Assuming you have created a warped binary image called "binary_warped"
# Take a histogram of the bottom half of the image
histogram = np.sum(binary_warped[binary_warped.shape[0]//2:,:], axis=0)
# Create an output image to draw on and visualize the result
out_img = np.dstack((binary_warped, binary_warped, binary_warped))*255
# Find the peak of the left and right halves of the histogram
# These will be the starting point for the left and right lines
midpoint = np.int(histogram.shape[0]/2)
leftx_base = np.argmax(histogram[:midpoint])
rightx_base = np.argmax(histogram[midpoint:]) + midpoint
# Choose the number of sliding windows
nwindows = 10
# Set height of windows
window_height = np.int(binary_warped.shape[0]/nwindows)
# Identify the x and y positions of all nonzero pixels in the image
nonzero = binary_warped.nonzero()
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Current positions to be updated for each window
leftx_current = leftx_base
rightx_current = rightx_base
# Set the width of the windows +/- margin
margin = 80
# Set minimum number of pixels found to recenter window
minpix = 50
# Create empty lists to receive left and right lane pixel indices
left_lane_inds = []
right_lane_inds = []
# Step through the windows one by one
for window in range(nwindows):
# Identify window boundaries in x and y (and right and left)
win_y_low = binary_warped.shape[0] - (window+1)*window_height
win_y_high = binary_warped.shape[0] - window*window_height
win_xleft_low = leftx_current - margin
win_xleft_high = leftx_current + margin
win_xright_low = rightx_current - margin
win_xright_high = rightx_current + margin
# Draw the windows on the visualization image
cv2.rectangle(out_img,(win_xleft_low,win_y_low),(win_xleft_high,win_y_high),
(0,255,0), 2)
cv2.rectangle(out_img,(win_xright_low,win_y_low),(win_xright_high,win_y_high),
(0,255,0), 2)
# Identify the nonzero pixels in x and y within the window
good_left_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xleft_low) & (nonzerox < win_xleft_high)).nonzero()[0]
good_right_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xright_low) & (nonzerox < win_xright_high)).nonzero()[0]
# Append these indices to the lists
left_lane_inds.append(good_left_inds)
right_lane_inds.append(good_right_inds)
# If you found > minpix pixels, recenter next window on their mean position
if len(good_left_inds) > minpix:
leftx_current = np.int(np.mean(nonzerox[good_left_inds]))
if len(good_right_inds) > minpix:
rightx_current = np.int(np.mean(nonzerox[good_right_inds]))
# Concatenate the arrays of indices
left_lane_inds = np.concatenate(left_lane_inds)
right_lane_inds = np.concatenate(right_lane_inds)
# Extract left and right line pixel positions
leftx = nonzerox[left_lane_inds]
lefty = nonzeroy[left_lane_inds]
rightx = nonzerox[right_lane_inds]
righty = nonzeroy[right_lane_inds]
# Fit a second order polynomial to each
left_fit = np.polyfit(lefty, leftx, 2)
right_fit = np.polyfit(righty, rightx, 2)
In [ ]:
In [8]:
def slide_window(img):
masked, binary_warped, color_binary, s_binary, sxbinary,Minv = threshold(img)
# Assuming you have created a warped binary image called "binary_warped"
# Take a histogram of the bottom half of the image
histogram = np.sum(binary_warped[binary_warped.shape[0]//2:,:], axis=0)
# Create an output image to draw on and visualize the result
out_img = np.dstack((binary_warped, binary_warped, binary_warped))*255
# Find the peak of the left and right halves of the histogram
# These will be the starting point for the left and right lines
midpoint = np.int(histogram.shape[0]/2)
leftx_base = np.argmax(histogram[:midpoint])
rightx_base = np.argmax(histogram[midpoint:]) + midpoint
# Choose the number of sliding windows
nwindows = 10
# Set height of windows
window_height = np.int(binary_warped.shape[0]/nwindows)
# Identify the x and y positions of all nonzero pixels in the image
nonzero = binary_warped.nonzero()
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Current positions to be updated for each window
leftx_current = leftx_base
rightx_current = rightx_base
# Set the width of the windows +/- margin
margin = 80
# Set minimum number of pixels found to recenter window
minpix = 50
# Create empty lists to receive left and right lane pixel indices
left_lane_inds = []
right_lane_inds = []
# Step through the windows one by one
for window in range(nwindows):
# Identify window boundaries in x and y (and right and left)
win_y_low = binary_warped.shape[0] - (window+1)*window_height
win_y_high = binary_warped.shape[0] - window*window_height
win_xleft_low = leftx_current - margin
win_xleft_high = leftx_current + margin
win_xright_low = rightx_current - margin
win_xright_high = rightx_current + margin
# Draw the windows on the visualization image
cv2.rectangle(out_img,(win_xleft_low,win_y_low),(win_xleft_high,win_y_high),
(0,255,0), 2)
cv2.rectangle(out_img,(win_xright_low,win_y_low),(win_xright_high,win_y_high),
(0,255,0), 2)
# Identify the nonzero pixels in x and y within the window
good_left_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xleft_low) & (nonzerox < win_xleft_high)).nonzero()[0]
good_right_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xright_low) & (nonzerox < win_xright_high)).nonzero()[0]
# Append these indices to the lists
left_lane_inds.append(good_left_inds)
right_lane_inds.append(good_right_inds)
# If you found > minpix pixels, recenter next window on their mean position
if len(good_left_inds) > minpix:
leftx_current = np.int(np.mean(nonzerox[good_left_inds]))
if len(good_right_inds) > minpix:
rightx_current = np.int(np.mean(nonzerox[good_right_inds]))
# Concatenate the arrays of indices
left_lane_inds = np.concatenate(left_lane_inds)
right_lane_inds = np.concatenate(right_lane_inds)
# Extract left and right line pixel positions
leftx = nonzerox[left_lane_inds]
lefty = nonzeroy[left_lane_inds]
rightx = nonzerox[right_lane_inds]
righty = nonzeroy[right_lane_inds]
# Fit a second order polynomial to each
left_fit = np.polyfit(lefty, leftx, 2)
right_fit = np.polyfit(righty, rightx, 2)
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
out_img[nonzeroy[left_lane_inds], nonzerox[left_lane_inds]] = [255, 0, 0]
out_img[nonzeroy[right_lane_inds], nonzerox[right_lane_inds]] = [0, 0, 255]
# Generate x and y values for plotting
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
leftx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
rightx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
leftx = leftx[::-1] # Reverse to match top-to-bottom in y
rightx = rightx[::-1] # Reverse to match top-to-bottom in y
# Fit a second order polynomial to pixel positions in each fake lane line
left_fit = np.polyfit(ploty, leftx, 2)
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fit = np.polyfit(ploty, rightx, 2)
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
leftx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
rightx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
leftx = leftx[::-1] # Reverse to match top-to-bottom in y
rightx = rightx[::-1] # Reverse to match top-to-bottom in y
# Fit a second order polynomial to pixel positions in each fake lane line
left_fit = np.polyfit(ploty, leftx, 2)
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fit = np.polyfit(ploty, rightx, 2)
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
return rightx, leftx, ploty, left_fit, right_fit, binary_warped, Minv
In [87]:
# Generate x and y values for plotting
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
out_img[nonzeroy[left_lane_inds], nonzerox[left_lane_inds]] = [255, 0, 0]
out_img[nonzeroy[right_lane_inds], nonzerox[right_lane_inds]] = [0, 0, 255]
plt.imshow(out_img)
plt.plot(left_fitx, ploty, color='yellow')
plt.plot(right_fitx, ploty, color='yellow')
plt.xlim(0, 1280)
plt.ylim(720, 0)
Out[87]:
In [90]:
# Generate some fake data to represent lane-line pixels
#ploty = np.linspace(0, 719, num=720)# to cover same y-range as image
#quadratic_coeff = 3e-4 # arbitrary quadratic coefficient
# For each y position generate random x position within +/-50 pix
# of the line base position in each case (x=200 for left, and x=900 for right)
#leftx = np.array([200 + (y**2)*quadratic_coeff + np.random.randint(-50, high=51)
# for y in ploty])
#rightx = np.array([900 + (y**2)*quadratic_coeff + np.random.randint(-50, high=51)
# for y in ploty])
# Generate x and y values for plotting
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
leftx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
rightx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
leftx = leftx[::-1] # Reverse to match top-to-bottom in y
rightx = rightx[::-1] # Reverse to match top-to-bottom in y
# Fit a second order polynomial to pixel positions in each fake lane line
left_fit = np.polyfit(ploty, leftx, 2)
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fit = np.polyfit(ploty, rightx, 2)
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
# Plot up the fake data
mark_size = 3
plt.plot(leftx, ploty, 'o', color='red', markersize=mark_size)
plt.plot(rightx, ploty, 'o', color='blue', markersize=mark_size)
plt.xlim(0, 1280)
plt.ylim(0, 720)
plt.plot(left_fitx, ploty, color='green', linewidth=3)
plt.plot(right_fitx, ploty, color='green', linewidth=3)
plt.gca().invert_yaxis() # to visualize as we do the images
Measuring Curvature¶
In [91]:
# Define y-value where we want radius of curvature
# I'll choose the maximum y-value, corresponding to the bottom of the image
y_eval = np.max(ploty)
left_curverad = ((1 + (2*left_fit[0]*y_eval + left_fit[1])**2)**1.5) / np.absolute(2*left_fit[0])
right_curverad = ((1 + (2*right_fit[0]*y_eval + right_fit[1])**2)**1.5) / np.absolute(2*right_fit[0])
print(left_curverad, right_curverad)
In [92]:
# Define conversions in x and y from pixels space to meters
ym_per_pix = 30/720 # meters per pixel in y dimension
xm_per_pix = 3.7/700 # meters per pixel in x dimension
# Fit new polynomials to x,y in world space
left_fit_cr = np.polyfit(ploty*ym_per_pix, leftx*xm_per_pix, 2)
right_fit_cr = np.polyfit(ploty*ym_per_pix, rightx*xm_per_pix, 2)
# Calculate the new radii of curvature
left_curverad = ((1 + (2*left_fit_cr[0]*y_eval*ym_per_pix + left_fit_cr[1])**2)**1.5) / np.absolute(2*left_fit_cr[0])
right_curverad = ((1 + (2*right_fit_cr[0]*y_eval*ym_per_pix + right_fit_cr[1])**2)**1.5) / np.absolute(2*right_fit_cr[0])
# Now our radius of curvature is in meters
print(left_curverad, 'm', right_curverad, 'm')
# Example values: 632.1 m 626.2 m
In [93]:
left_lane_bottom = (left_fit[0]*y_eval)**2 + left_fit[0]*y_eval + left_fit[2]
right_lane_bottom = (right_fit[0]*y_eval)**2 + right_fit[0]*y_eval + right_fit[2]
# Lane center as mid of left and right lane bottom
lane_center = (left_lane_bottom + right_lane_bottom)/2.
center_image = 640
center = (lane_center - center_image)*xm_per_pix #Convert to meters
print(left_curverad, 'm', right_curverad, 'm', center, 'm')
In [9]:
def curvature(left_fit, right_fit, rightx, leftx, ploty):
# Define y-value where we want radius of curvature
# I'll choose the maximum y-value, corresponding to the bottom of the image
y_eval = np.max(ploty)
left_curverad = ((1 + (2*left_fit[0]*y_eval + left_fit[1])**2)**1.5) / np.absolute(2*left_fit[0])
right_curverad = ((1 + (2*right_fit[0]*y_eval + right_fit[1])**2)**1.5) / np.absolute(2*right_fit[0])
# Define conversions in x and y from pixels space to meters
ym_per_pix = 30/720 # meters per pixel in y dimension
xm_per_pix = 3.7/700 # meters per pixel in x dimension
# Fit new polynomials to x,y in world space
left_fit_cr = np.polyfit(ploty*ym_per_pix, leftx*xm_per_pix, 2)
right_fit_cr = np.polyfit(ploty*ym_per_pix, rightx*xm_per_pix, 2)
# Calculate the new radii of curvature
left_curverad = ((1 + (2*left_fit_cr[0]*y_eval*ym_per_pix + left_fit_cr[1])**2)**1.5) / np.absolute(2*left_fit_cr[0])
right_curverad = ((1 + (2*right_fit_cr[0]*y_eval*ym_per_pix + right_fit_cr[1])**2)**1.5) / np.absolute(2*right_fit_cr[0])
# Now our radius of curvature is in meters
left_lane_bottom = (left_fit[0]*y_eval)**2 + left_fit[0]*y_eval + left_fit[2]
right_lane_bottom = (right_fit[0]*y_eval)**2 + right_fit[0]*y_eval + right_fit[2]
# Lane center as mid of left and right lane bottom
lane_center = (left_lane_bottom + right_lane_bottom)/2.
center_image = 640
center = (lane_center - center_image)*xm_per_pix #Convert to meters
return left_curverad, right_curverad, center
Draw lane¶
In [10]:
def draw_lines(undist, warped,left_fit, right_fit, left_cur, right_cur, center, Minv, show_img = True ):
# Create an image to draw the lines on
warp_zero = np.zeros_like(warped).astype(np.uint8)
color_warp = np.dstack((warp_zero, warp_zero, warp_zero))
ploty = np.linspace(0, warped.shape[0]-1, warped.shape[0] )
# Fit new polynomials to x,y in world space
left_fitx = left_fit[0]*ploty**2+left_fit[1]*ploty+left_fit[2]
right_fitx = right_fit[0]*ploty**2+right_fit[1]*ploty+right_fit[2]
# Recast the x and y points into usable format for cv2.fillPoly()
pts_left = np.array([np.transpose(np.vstack([left_fitx, ploty]))])
pts_right = np.array([np.flipud(np.transpose(np.vstack([right_fitx, ploty])))])
pts = np.hstack((pts_left, pts_right))
# Draw the lane onto the warped blank image
cv2.fillPoly(color_warp, np.int_([pts]), (0,255, 0))
# Warp the blank back to original image space using inverse perspective matrix (Minv)
newwarp = cv2.warpPerspective(color_warp, Minv, (color_warp.shape[1], color_warp.shape[0]))
# Combine the result with the original image
result = cv2.addWeighted(undist, 1, newwarp, 0.3, 0)
add_text_to_image(result, left_cur, right_cur, center)
if show_img == True:
plt.figure(figsize=(20,20))
fig = plt.figure()
plt.imshow(cv2.cvtColor(result, cv2.COLOR_BGR2RGB))
return result
In [11]:
def add_text_to_image(img, left_cur, right_cur, center):
"""
Draws information about the center offset and the current lane curvature onto the given image.
:param img:
"""
cur = (left_cur + right_cur)/2.
font = cv2.FONT_HERSHEY_SIMPLEX
cv2.putText(img, 'Radius of Curvature = %d(m)' % cur, (50, 50), font, 1, (255, 255, 255), 2)
left_or_right = "left" if center < 0 else "right"
cv2.putText(img, 'Vehicle is %.2fm %s of center' % (np.abs(center), left_or_right), (50, 100), font, 1,
(255, 255, 255), 2)
In [104]:
img = cv2.imread("test_images/straight_lines1.jpg")
result = draw_lines(undistort(img),binary_warped,left_fit, right_fit, left_curverad, right_curverad, center, show_img = True )
In [107]:
img = cv2.imread("test_images/straight_lines1.jpg")
leftx, ploty, left_fit, right_fit, binary_warped = slide_window(img)
left_curverad, right_curverad, center = curvature(left_fit, right_fit, leftx, ploty)
result = draw_lines(undistort(img),binary_warped,left_fit, right_fit, left_curverad, right_curverad, center, show_img = True )
In [12]:
images = glob.glob('test_images/*.jpg')
for image in images:
img = cv2.imread(image)
rightx, leftx, ploty, left_fit, right_fit, binary_warped, Minv= slide_window(img)
left_curverad, right_curverad, center= curvature(left_fit, right_fit, rightx, leftx,ploty)
print((left_curverad+right_curverad)/2)
result = draw_lines(undistort(img),binary_warped,left_fit, right_fit, left_curverad, right_curverad, center, Minv, show_img = False )
Test on Videos¶
In [11]:
# Import everything needed to edit/save/watch video clips
from moviepy.editor import VideoFileClip
from IPython.display import HTML
Full pipline¶
In [16]:
#importing some useful packages
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import numpy as np
import cv2
import glob
%matplotlib inline
objpoints = [] # 3d points in real world space
imgpoints = [] # 2d points in image plane.
#Prepare object points, like(0,0,0), (1,0,0)...
objp = np.zeros((6*9,3), np.float32)
objp[:,:2] = np.mgrid[0:9, 0:6].T.reshape(-1,2) #x, y coordinates
images = glob.glob('camera_cal/calibration*.jpg') # all images path
for i, filename in enumerate(images):
img = cv2.imread(filename)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# Find the chessboard corners
ret, corners = cv2.findChessboardCorners(gray, (9,6), None)
# If found, add object points, image points
if ret == True:
objpoints.append(objp)
imgpoints.append(corners)
# define undistort function
def undistort(img):
img_size = (img.shape[1], img.shape[0])
ret, mtx, dist, rvecs, tvecs = cv2.calibrateCamera(objpoints, imgpoints, img_size, None, None)
undst = cv2.undistort(img, mtx, dist, None, mtx)
return undst
#define perspective transform function
def per_transform(undist):
img_size = (undist.shape[1], undist.shape[0])
src = np.float32([[220,720],[1110,720],[722,470],[570,470]])
dst = np.float32([[320,720],[920,720],[920,1],[320,1]])
M = cv2.getPerspectiveTransform(src, dst)
Minv = cv2.getPerspectiveTransform(dst, src)
warped = cv2.warpPerspective(img, M, img_size, flags=cv2.INTER_LINEAR)
unpersp = cv2.warpPerspective(warped, Minv, img_size, flags=cv2.INTER_LINEAR)
return warped, unpersp, Minv
#Define a function for lane detection
def threshold(img):
#undistort image
undist = undistort(img)
#per-transform
wraped, unpersp, Minv = per_transform(undist)
#graysclae
gray = cv2.cvtColor(wraped, cv2.COLOR_BGR2GRAY)
#HLS color space
hls = cv2.cvtColor(wraped, cv2.COLOR_BGR2HLS)
s_channel = hls[:,:,2]
# apply gradient threshold on the horizontal gradient
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0) # Take the derivative in x
abs_sobelx = np.absolute(sobelx) # Absolute x derivative to accentuate lines away from horizontal
scaled_sobel = np.uint8(255*abs_sobelx/np.max(abs_sobelx))
# Threshold color channel
thresh_min = 20
thresh_max = 100
sxbinary = np.zeros_like(scaled_sobel)
sxbinary[(scaled_sobel >= thresh_min) & (scaled_sobel <= thresh_max)] = 1
# Threshold color channel
s_thresh_min = 90
s_thresh_max = 255
s_binary = np.zeros_like(s_channel)
s_binary[(s_channel >= s_thresh_min) & (s_channel <= s_thresh_max)] = 1
# Stack each channel to view their individual contributions in green and blue respectively
# This returns a stack of the two binary images, whose components you can see as different colors
color_binary = np.dstack(( np.zeros_like(sxbinary), sxbinary, s_binary))
# Combine the two binary thresholds
combined_binary = np.zeros_like(sxbinary)
combined_binary[(s_binary == 1) | (sxbinary == 1)] = 1
#region of interst
mask = np.zeros_like(combined_binary)
length = mask.shape[1]
height = mask.shape[0]
vertices = np.array([[(0,height),(length/2, height/2), (length-1, height-1)]], dtype=np.int32)
cv2.fillPoly(mask, vertices, 1)
masked_image = cv2.bitwise_and(combined_binary, mask)
return masked_image, combined_binary, color_binary, s_binary, sxbinary, Minv
def slide_window(img):
masked, binary_warped, color_binary, s_binary, sxbinary,Minv = threshold(img)
# Assuming you have created a warped binary image called "binary_warped"
# Take a histogram of the bottom half of the image
histogram = np.sum(binary_warped[binary_warped.shape[0]//2:,:], axis=0)
# Create an output image to draw on and visualize the result
out_img = np.dstack((binary_warped, binary_warped, binary_warped))*255
# Find the peak of the left and right halves of the histogram
# These will be the starting point for the left and right lines
midpoint = np.int(histogram.shape[0]/2)
leftx_base = np.argmax(histogram[:midpoint])
rightx_base = np.argmax(histogram[midpoint:]) + midpoint
# Choose the number of sliding windows
nwindows = 10
# Set height of windows
window_height = np.int(binary_warped.shape[0]/nwindows)
# Identify the x and y positions of all nonzero pixels in the image
nonzero = binary_warped.nonzero()
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Current positions to be updated for each window
leftx_current = leftx_base
rightx_current = rightx_base
# Set the width of the windows +/- margin
margin = 80
# Set minimum number of pixels found to recenter window
minpix = 50
# Create empty lists to receive left and right lane pixel indices
left_lane_inds = []
right_lane_inds = []
# Step through the windows one by one
for window in range(nwindows):
# Identify window boundaries in x and y (and right and left)
win_y_low = binary_warped.shape[0] - (window+1)*window_height
win_y_high = binary_warped.shape[0] - window*window_height
win_xleft_low = leftx_current - margin
win_xleft_high = leftx_current + margin
win_xright_low = rightx_current - margin
win_xright_high = rightx_current + margin
# Draw the windows on the visualization image
cv2.rectangle(out_img,(win_xleft_low,win_y_low),(win_xleft_high,win_y_high),
(0,255,0), 2)
cv2.rectangle(out_img,(win_xright_low,win_y_low),(win_xright_high,win_y_high),
(0,255,0), 2)
# Identify the nonzero pixels in x and y within the window
good_left_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xleft_low) & (nonzerox < win_xleft_high)).nonzero()[0]
good_right_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xright_low) & (nonzerox < win_xright_high)).nonzero()[0]
# Append these indices to the lists
left_lane_inds.append(good_left_inds)
right_lane_inds.append(good_right_inds)
# If you found > minpix pixels, recenter next window on their mean position
if len(good_left_inds) > minpix:
leftx_current = np.int(np.mean(nonzerox[good_left_inds]))
if len(good_right_inds) > minpix:
rightx_current = np.int(np.mean(nonzerox[good_right_inds]))
# Concatenate the arrays of indices
left_lane_inds = np.concatenate(left_lane_inds)
right_lane_inds = np.concatenate(right_lane_inds)
# Extract left and right line pixel positions
leftx = nonzerox[left_lane_inds]
lefty = nonzeroy[left_lane_inds]
rightx = nonzerox[right_lane_inds]
righty = nonzeroy[right_lane_inds]
# Fit a second order polynomial to each
left_fit = np.polyfit(lefty, leftx, 2)
right_fit = np.polyfit(righty, rightx, 2)
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
out_img[nonzeroy[left_lane_inds], nonzerox[left_lane_inds]] = [255, 0, 0]
out_img[nonzeroy[right_lane_inds], nonzerox[right_lane_inds]] = [0, 0, 255]
# Generate x and y values for plotting
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
leftx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
rightx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
leftx = leftx[::-1] # Reverse to match top-to-bottom in y
rightx = rightx[::-1] # Reverse to match top-to-bottom in y
# Fit a second order polynomial to pixel positions in each fake lane line
left_fit = np.polyfit(ploty, leftx, 2)
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fit = np.polyfit(ploty, rightx, 2)
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
leftx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
rightx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
leftx = leftx[::-1] # Reverse to match top-to-bottom in y
rightx = rightx[::-1] # Reverse to match top-to-bottom in y
# Fit a second order polynomial to pixel positions in each fake lane line
left_fit = np.polyfit(ploty, leftx, 2)
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fit = np.polyfit(ploty, rightx, 2)
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
return rightx, leftx, ploty, left_fit, right_fit, binary_warped, Minv
def curvature(left_fit, right_fit, rightx, leftx, ploty):
# Define y-value where we want radius of curvature
# I'll choose the maximum y-value, corresponding to the bottom of the image
y_eval = np.max(ploty)
left_curverad = ((1 + (2*left_fit[0]*y_eval + left_fit[1])**2)**1.5) / np.absolute(2*left_fit[0])
right_curverad = ((1 + (2*right_fit[0]*y_eval + right_fit[1])**2)**1.5) / np.absolute(2*right_fit[0])
# Define conversions in x and y from pixels space to meters
ym_per_pix = 30/720 # meters per pixel in y dimension
xm_per_pix = 3.7/700 # meters per pixel in x dimension
# Fit new polynomials to x,y in world space
left_fit_cr = np.polyfit(ploty*ym_per_pix, leftx*xm_per_pix, 2)
right_fit_cr = np.polyfit(ploty*ym_per_pix, rightx*xm_per_pix, 2)
# Calculate the new radii of curvature
left_curverad = ((1 + (2*left_fit_cr[0]*y_eval*ym_per_pix + left_fit_cr[1])**2)**1.5) / np.absolute(2*left_fit_cr[0])
right_curverad = ((1 + (2*right_fit_cr[0]*y_eval*ym_per_pix + right_fit_cr[1])**2)**1.5) / np.absolute(2*right_fit_cr[0])
# Now our radius of curvature is in meters
left_lane_bottom = (left_fit[0]*y_eval)**2 + left_fit[0]*y_eval + left_fit[2]
right_lane_bottom = (right_fit[0]*y_eval)**2 + right_fit[0]*y_eval + right_fit[2]
# Lane center as mid of left and right lane bottom
lane_center = (left_lane_bottom + right_lane_bottom)/2.
center_image = 640
center = (lane_center - center_image)*xm_per_pix #Convert to meters
return left_curverad, right_curverad, center
def draw_lines(undist, warped,left_fit, right_fit, left_cur, right_cur, center, Minv, show_img = True ):
# Create an image to draw the lines on
warp_zero = np.zeros_like(warped).astype(np.uint8)
color_warp = np.dstack((warp_zero, warp_zero, warp_zero))
ploty = np.linspace(0, warped.shape[0]-1, warped.shape[0] )
# Fit new polynomials to x,y in world space
left_fitx = left_fit[0]*ploty**2+left_fit[1]*ploty+left_fit[2]
right_fitx = right_fit[0]*ploty**2+right_fit[1]*ploty+right_fit[2]
# Recast the x and y points into usable format for cv2.fillPoly()
pts_left = np.array([np.transpose(np.vstack([left_fitx, ploty]))])
pts_right = np.array([np.flipud(np.transpose(np.vstack([right_fitx, ploty])))])
pts = np.hstack((pts_left, pts_right))
# Draw the lane onto the warped blank image
cv2.fillPoly(color_warp, np.int_([pts]), (0,255, 0))
# Warp the blank back to original image space using inverse perspective matrix (Minv)
newwarp = cv2.warpPerspective(color_warp, Minv, (color_warp.shape[1], color_warp.shape[0]))
# Combine the result with the original image
result = cv2.addWeighted(undist, 1, newwarp, 0.3, 0)
add_text_to_image(result, left_cur, right_cur, center)
if show_img == True:
plt.figure(figsize=(20,20))
fig = plt.figure()
plt.imshow(cv2.cvtColor(result, cv2.COLOR_BGR2RGB))
return result
def add_text_to_image(img, left_cur, right_cur, center):
"""
Draws information about the center offset and the current lane curvature onto the given image.
:param img:
"""
cur = (left_cur + right_cur)/2.
font = cv2.FONT_HERSHEY_SIMPLEX
cv2.putText(img, 'Radius of Curvature = %d(m)' % cur, (50, 50), font, 1, (255, 255, 255), 2)
left_or_right = "left" if center < 0 else "right"
cv2.putText(img, 'Vehicle is %.2fm %s of center' % (np.abs(center), left_or_right), (50, 100), font, 1,
(255, 255, 255), 2)
In [17]:
def process_image(image):
rightx, leftx, ploty, left_fit, right_fit, binary_warped, Minv= slide_window(img)
left_curverad, right_curverad, center= curvature(left_fit, right_fit, rightx, leftx,ploty)
#print((left_curverad+right_curverad)/2)
result = draw_lines(undistort(img),binary_warped,left_fit, right_fit, left_curverad, right_curverad, center, Minv, show_img = False)
result = cv2.cvtColor(result, cv2.COLOR_BGR2RGB)
return result
In [18]:
images = glob.glob('test_images/*.jpg')
for image in images:
img = cv2.imread(image)
result = process_image(img)
plt.imshow(result)
plt.show()
In [13]:
white_output = 'project_video_output.mp4'
## To speed up the testing process you may want to try your pipeline on a shorter subclip of the video
## To do so add .subclip(start_second,end_second) to the end of the line below
## Where start_second and end_second are integer values representing the start and end of the subclip
## You may also uncomment the following line for a subclip of the first 5 seconds
clip1 = VideoFileClip("project_video.mp4").subclip(0,1)
#clip1 = VideoFileClip("project_video.mp4")
white_clip = clip1.fl_image(process_image) #NOTE: this function expects color images!!
%time white_clip.write_videofile(white_output, audio=False)
Play the video inline, or if you prefer find the video in your filesystem (should be in the same directory) and play it in your video player of choice.
In [14]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(white_output))
Out[14]:
In [ ]: